Tessera adapted “jumping” genetic elements to insert a gene into T-cells inside mice: two RNAs delivered in nanoparticles generated CAR-T-cells that killed B-cells, which remained depleted throughout the 38-day experiment

Tessera adapted “jumping” genetic elements to insert a gene into T-cells inside mice: two RNAs delivered in nanoparticles generated CAR-T-cells that killed B-cells, which remained depleted throughout the 38-day experiment

On October 21, researchers at Tessera Therapeutics published a paper in Cell. CAR-T-cells are T-cells equipped with an engineered receptor that helps them recognize and kill target cells. The experiment used immunodeficient mice transplanted with human blood stem cells, giving them human T-cells and B-cells.

CAR-T-cells are prepared outside the body: T-cells are collected from a patient, a lentiviral vector (a virus used as a delivery vehicle) inserts the receptor gene into their chromosomes, and the cells are returned to the patient. A viral vector administered directly into the body produces a permanent insertion but is difficult to manufacture. RNA delivered in lipid nanoparticles is simpler to use but breaks down quickly. The authors aim to combine the stability of viral gene insertion with the simplicity of RNA delivery.

A retrotransposon is a “jumping” genetic element that can copy itself. The cell transcribes it into RNA and uses that RNA to make a protein. The protein then nicks a chromosome and copies the same RNA back into DNA at the nick. It normally copies only its own RNA. Tessera split the system into two RNAs: an mRNA that encodes the protein and a template RNA carrying the desired gene between two terminal sequences. The authors propose that these terminal sequences recruit the matching protein to the template. They named these retrotransposons “Gene Writers.”

RNA systems based on R2 retrotransposons, described in Cell and Nature Biotechnology, direct insertion into repeated ribosomal RNA genes. Typewriter, a startup with $56 million in funding, is also using R2 to generate CAR-T-cells inside the body and considers these genes safe insertion sites. The paper’s authors view this targeting as a possible limitation. Tessera therefore screened 9 559 retroelements from different species and selected 406 candidates with terminal sequences. Only ten inserted a fluorescent reporter gene into human cells, and the four most active in T-cells worked only with a template whose terminal sequences matched the protein. For T-cells, the researchers chose a Gene Writer from the genome of an anole lizard because of its high activity.

In cultured T-cells, Gene Writers insert genes at random sites in the genome, whereas lentiviral insertion favors genes and risks damaging them. Gene Writer insertions are usually truncated: the gene encoding a receptor for the B-cell protein CD19 remains intact in 22,8% of insertions, compared with 39% for the shorter reporter gene. An extra or missing chromosome was found in 1% of cells after Gene Writer treatment, the same proportion as in untreated cells, and in 8–10% after editing two genes with CRISPR-Cas9.

In mice, the receptor targeted CD20, another B-cell protein, so B-cell depletion served as a measure of CAR-T-cell activity. Six mice received a single dose of nanoparticles containing the two RNAs and antibodies against CD3, a T-cell protein. On day 11, CAR-T-cells accounted for about 15% of T-cells. By day 18, B-cells were almost completely depleted, and they remained undetectable in the blood and spleen through the end of the experiment on day 38. For comparison, four mice received receptor mRNA in nanoparticles bearing antibodies against CD8, which produced greater B-cell depletion than antibodies against CD3. B-cell numbers fell substantially by day 3 but recovered by day 38.

In October 2024, Jeremy Duffield, then chief scientific officer at Prime Medicine, compared Tessera’s platform as a whole with his company’s method, prime editing: “[It is] a Cas domain attached to a reverse transcriptase, so it is probably quite similar.” The Gene Writer in this paper is a single retrotransposon protein that both nicks DNA, as a Cas domain does, and copies RNA into DNA, as a reverse transcriptase does. Cas9 was attached to it only in an experiment aimed at directing insertion to a chosen site, which achieved low efficiency.